Search

Your search keyword '"Li, Youhua"' showing total 174 results
Start Over Author "Li, Youhua"
174 results on '"Li, Youhua"'

1. An Empirical Study of Training ID-Agnostic Multi-modal Sequential Recommenders

Author

Li, Youhua, Du, Hanwen, Ni, Yongxin, He, Yuanqi, Fu, Junchen, Liu, Xiangyan, and Guo, Qi

Subjects
Computer Science - Information Retrieval
Abstract

Sequential Recommendation (SR) aims to predict future user-item interactions based on historical interactions. While many SR approaches concentrate on user IDs and item IDs, the human perception of the world through multi-modal signals, like text and images, has inspired researchers to delve into constructing SR from multi-modal information without using IDs. However, the complexity of multi-modal learning manifests in diverse feature extractors, fusion methods, and pre-trained models. Consequently, designing a simple and universal \textbf{M}ulti-\textbf{M}odal \textbf{S}equential \textbf{R}ecommendation (\textbf{MMSR}) framework remains a formidable challenge. We systematically summarize the existing multi-modal related SR methods and distill the essence into four core components: visual encoder, text encoder, multimodal fusion module, and sequential architecture. Along these dimensions, we dissect the model designs, and answer the following sub-questions: First, we explore how to construct MMSR from scratch, ensuring its performance either on par with or exceeds existing SR methods without complex techniques. Second, we examine if MMSR can benefit from existing multi-modal pre-training paradigms. Third, we assess MMSR's capability in tackling common challenges like cold start and domain transferring. Our experiment results across four real-world recommendation scenarios demonstrate the great potential ID-agnostic multi-modal sequential recommendation. Our framework can be found at: https://github.com/MMSR23/MMSR., Comment: We are requesting to withdraw the paper due to a significant methodological error discovered in the experimental setup, specifically in Section 4.3. This error affects the validity of the results and conclusions drawn from the study. We intend to address these issues and submit a corrected version in the future

Published
2024

2. Multi-Modality is All You Need for Transferable Recommender Systems

Author

Li, Youhua, Du, Hanwen, Ni, Yongxin, Zhao, Pengpeng, Guo, Qi, Yuan, Fajie, and Zhou, Xiaofang

Subjects
Computer Science - Information Retrieval
Abstract

ID-based Recommender Systems (RecSys), where each item is assigned a unique identifier and subsequently converted into an embedding vector, have dominated the designing of RecSys. Though prevalent, such ID-based paradigm is not suitable for developing transferable RecSys and is also susceptible to the cold-start issue. In this paper, we unleash the boundaries of the ID-based paradigm and propose a Pure Multi-Modality based Recommender system (PMMRec), which relies solely on the multi-modal contents of the items (e.g., texts and images) and learns transition patterns general enough to transfer across domains and platforms. Specifically, we design a plug-and-play framework architecture consisting of multi-modal item encoders, a fusion module, and a user encoder. To align the cross-modal item representations, we propose a novel next-item enhanced cross-modal contrastive learning objective, which is equipped with both inter- and intra-modality negative samples and explicitly incorporates the transition patterns of user behaviors into the item encoders. To ensure the robustness of user representations, we propose a novel noised item detection objective and a robustness-aware contrastive learning objective, which work together to denoise user sequences in a self-supervised manner. PMMRec is designed to be loosely coupled, so after being pre-trained on the source data, each component can be transferred alone, or in conjunction with other components, allowing PMMRec to achieve versatility under both multi-modality and single-modality transfer learning settings. Extensive experiments on 4 sources and 10 target datasets demonstrate that PMMRec surpasses the state-of-the-art recommenders in both recommendation performance and transferability. Our code and dataset is available at: https://github.com/ICDE24/PMMRec., Comment: ICDE'24 Accepted

Published
2023

4. A Content-Driven Micro-Video Recommendation Dataset at Scale

Author

Ni, Yongxin, Cheng, Yu, Liu, Xiangyan, Fu, Junchen, Li, Youhua, He, Xiangnan, Zhang, Yongfeng, and Yuan, Fajie

Subjects
Computer Science - Information Retrieval
Abstract

Micro-videos have recently gained immense popularity, sparking critical research in micro-video recommendation with significant implications for the entertainment, advertising, and e-commerce industries. However, the lack of large-scale public micro-video datasets poses a major challenge for developing effective recommender systems. To address this challenge, we introduce a very large micro-video recommendation dataset, named "MicroLens", consisting of one billion user-item interaction behaviors, 34 million users, and one million micro-videos. This dataset also contains various raw modality information about videos, including titles, cover images, audio, and full-length videos. MicroLens serves as a benchmark for content-driven micro-video recommendation, enabling researchers to utilize various modalities of video information for recommendation, rather than relying solely on item IDs or off-the-shelf video features extracted from a pre-trained network. Our benchmarking of multiple recommender models and video encoders on MicroLens has yielded valuable insights into the performance of micro-video recommendation. We believe that this dataset will not only benefit the recommender system community but also promote the development of the video understanding field. Our datasets and code are available at https://github.com/westlake-repl/MicroLens.

Published
2023

5. NineRec: A Benchmark Dataset Suite for Evaluating Transferable Recommendation

Author

Zhang, Jiaqi, Cheng, Yu, Ni, Yongxin, Pan, Yunzhu, Yuan, Zheng, Fu, Junchen, Li, Youhua, Wang, Jie, and Yuan, Fajie

Subjects
Computer Science - Information Retrieval
Abstract

Large foundational models, through upstream pre-training and downstream fine-tuning, have achieved immense success in the broad AI community due to improved model performance and significant reductions in repetitive engineering. By contrast, the transferable one-for-all models in the recommender system field, referred to as TransRec, have made limited progress. The development of TransRec has encountered multiple challenges, among which the lack of large-scale, high-quality transfer learning recommendation dataset and benchmark suites is one of the biggest obstacles. To this end, we introduce NineRec, a TransRec dataset suite that comprises a large-scale source domain recommendation dataset and nine diverse target domain recommendation datasets. Each item in NineRec is accompanied by a descriptive text and a high-resolution cover image. Leveraging NineRec, we enable the implementation of TransRec models by learning from raw multimodal features instead of relying solely on pre-extracted off-the-shelf features. Finally, we present robust TransRec benchmark results with several classical network architectures, providing valuable insights into the field. To facilitate further research, we will release our code, datasets, benchmarks, and leaderboards at https://github.com/westlake-repl/NineRec.

Published
2023

7. Where to Go Next for Recommender Systems? ID- vs. Modality-based Recommender Models Revisited

Author

Yuan, Zheng, Yuan, Fajie, Song, Yu, Li, Youhua, Fu, Junchen, Yang, Fei, Pan, Yunzhu, and Ni, Yongxin

Subjects
Computer Science - Information Retrieval
Abstract

Recommendation models that utilize unique identities (IDs) to represent distinct users and items have been state-of-the-art (SOTA) and dominated the recommender systems (RS) literature for over a decade. Meanwhile, the pre-trained modality encoders, such as BERT and ViT, have become increasingly powerful in modeling the raw modality features of an item, such as text and images. Given this, a natural question arises: can a purely modality-based recommendation model (MoRec) outperforms or matches a pure ID-based model (IDRec) by replacing the itemID embedding with a SOTA modality encoder? In fact, this question was answered ten years ago when IDRec beats MoRec by a strong margin in both recommendation accuracy and efficiency. We aim to revisit this `old' question and systematically study MoRec from several aspects. Specifically, we study several sub-questions: (i) which recommendation paradigm, MoRec or IDRec, performs better in practical scenarios, especially in the general setting and warm item scenarios where IDRec has a strong advantage? does this hold for items with different modality features? (ii) can the latest technical advances from other communities (i.e., natural language processing and computer vision) translate into accuracy improvement for MoRec? (iii) how to effectively utilize item modality representation, can we use it directly or do we have to adjust it with new data? (iv) are there some key challenges for MoRec to be solved in practical applications? To answer them, we conduct rigorous experiments for item recommendations with two popular modalities, i.e., text and vision. We provide the first empirical evidence that MoRec is already comparable to its IDRec counterpart with an expensive end-to-end training method, even for warm item recommendation. Our results potentially imply that the dominance of IDRec in the RS field may be greatly challenged in the future.

Published
2023

9. Study on the Root Crack Propagation Characteristics and Residual Life of Case-hardened Gears

Author

Wu Luji, Yang Shihao, Yang Linjie, and Li Youhua

Subjects
Case-hardened gear, Root crack, Stress intensity factor, Residual life, Mechanical engineering and machinery, TJ1-1570
Abstract

In order to study the influence of tooth root cracks on the fatigue life of case-hardened gears, taking an involute case-hardened gear as the research object, the mechanism of the tooth root fatigue crack growth is analyzed based on the fracture mechanics method and fatigue crack growth theory. The residual life analysis model of the tooth root crack propagation for case-hardened gears is established, considering the influence of load, initial crack size and initial crack location, etc. The evolution of stress intensity factor and crack propagation mechanism at different root crack propagation stages is studied. According to the bending fatigue test data of an involute case-hardened gear pair, the calculation model is analyzed and verified, and the accuracy of the model is proved. The results indicate that, compared with type Ⅱ and type Ⅲ, type Ⅰ stress intensity factor is the highest, which gradually decreases from the tooth surface to the direction of crack depth. With the increase of factors such as load, crack length, crack width, and initial crack distance from the center position of the tooth width, the residual life of crack propagation decreases accordingly.

Published
2023
Full Text
View/download PDF

10. Investigation of microfriction properties of graphene/AlCoCrFeNi high entropy alloy.

Author

Li, Youhua, Ma, Qianli, Yu, Hechun, Zhang, Suxiang, Zhang, Guoqing, and Wang, Wenbo

Subjects
*NANOINDENTATION, *STRAIN hardening, *COMPOSITE coating, *GRAPHENE, *MOLECULAR dynamics, *DISLOCATION density
Abstract

Applying graphene (Gr) coatings to high-entropy alloys (HEA) is anticipated to enhance their tribological characteristics. The current understanding of the mechanism by which the Gr/HEA is enhanced at the atomic level is still limited. Molecular dynamics simulations revealed the mechanical behavior and strengthening mechanism of the Gr/AlCoCrFeNi HEA during nanoindentation and nanoscratch. The results demonstrate a substantial increase in the indentation hardness of the Gr/AlCoCrFeNi HEA by about 2.4 times. When Gr changed from a single layer to three layers, it further improved (3.2 times for a double layer and 3.9 times for three layers). At the same time, the friction coefficient is effectively reduced. Furthermore, the elevated in-plane stiffness of the Gr coating leads to an expansion of the effective loading area, resulting in increased Shockley dislocation and stair-rod dislocation density within the Gr/AlCoCrFeNi HEA, thereby amplifying the strain hardening effect and reducing subsurface damage. Qualitative experiments confirmed the excellent wear resistance of the Gr/HEA, and coating Gr increased the width of scratches, effectively confirming our simulation results. These findings provide valuable insights for the development and design of Gr/HEA composite coatings with enhanced mechanical properties. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

17. STUDY ON THE INFLUENCE OF THE THREE-DIMENSIONAL MORPHOLOGY OF THE GEAR PITTING PITS ON STRESS CONCENTRATION OF TOOTH SURFACE

Author

LI YouHua, CHEN Chao, YANG ShiHao, ZHANG ZhiHong, WANG Jun, and LIU ZhongMing

Subjects
Pitting, Gear, Tooth surface, Dimensional change, Stress concentration factor, Mechanical engineering and machinery, TJ1-1570, Materials of engineering and construction. Mechanics of materials, TA401-492
Abstract

In order to study the influence of the three-dimensional appearance of the gear pitting pits on the stress concentration of the tooth surface, the effect of pitting pits with different shapes on stress concentration factor of gear was analyzed by finite element method. By changing the length, width and depth of ellipsoid pitting pits, the influence of different length-width ratio and depth-width ratio on stress concentration factor was analyzed, and the approximate calculation model of stress concentration factor of ellipsoid pitting pits was established. According to the contact fatigue test data of a gear, the calculation model established by this method was analyzed and verified, and the influence of the change of pitting pit size on the stress concentration factor of the gear was further analyzed and discussed. The results show that: the stress concentration of ellipsoid pits on the tooth surface was the most obvious, followed by conical pits, pentagonal pits and triangular pits when the basic size of pitting pits was the same. The stress concentration factor increases with the increase of depth-width ratio and length-width ratio of ellipsoidal pit.

Published
2022
Full Text
View/download PDF

18. List of contributors

Author

Birrell, Simon Andrew, primary, Cox, Ian, additional, den Uijl, Ingrid, additional, Hasan, Narjis, additional, Hughes, Josie, additional, Li, Youhua, additional, Lida, Fumiya, additional, Lynch, Patrick Joseph, additional, Lynham, Paul, additional, Minaei, Negin, additional, Mottram, Toby, additional, Niu, Cong, additional, Quealy, Sinead, additional, Ross, Dave, additional, Schlingmann, Norbert, additional, Thomson, Willie, additional, and Yan, Xiu-Tian, additional

Published
2022
Full Text
View/download PDF

22. Research on Design and Optimization of Micro-Hole Aerostatic Bearing in Vacuum Environment.

Author

Fan, Guozhen, Li, Youhua, Li, Yuehua, Zang, Libin, Zhao, Ming, Li, Zhanxin, Yu, Hechun, Xu, Jialiang, Liang, Hongfei, Zhang, Guoqing, and Hou, Weijie

Subjects
AIR resistance, WEATHER, GENETIC algorithms, REMOTE sensing, REMOTE control
Abstract

Micro-hole aerostatic bearings are important components in micro-low-gravity simulation of aerospace equipment, and the accuracy of micro-low-gravity simulation tests is affected by them. In order to eliminate the influence of air resistance on the attitude control accuracy of remote sensing satellites and achieve high fidelity of micro-low-gravity simulation tests, in this study, a design and parameter optimization method was proposed for micro-hole aerostatic bearings for a vacuum environment. Firstly, the theoretical analysis was conducted to investigate the impact of various bearing parameters and external conditions on the bearing load capacity and mass flow. Subsequently, a function model describing the variation in bearing load capacity and mass flow with bearing parameters was obtained utilizing a BP neural network. The parameters of aerostatic bearings in a vacuum environment were optimized using the non-dominated sorting genetic algorithm (NSGA-II) with the objectives of maximizing the load capacity and minimizing the mass flow. Subsequently, experimental tests were conducted on the optimized bearings in both atmospheric and vacuum conditions to evaluate their load capacity and mass flow. The results show that in a vacuum environment, the load capacity and mass flow of aerostatic bearings are increased compared to those in standard atmospheric conditions. Furthermore, it has been determined that the optimal solution for the bearing's load capacity and mass flow occurs when the bearing has an orifice aperture of 0.1 mm, 36 holes, and an orifice distribution diameter of 38.83 mm. The corresponding load capacity and mass flow are 460.644 N and 11.816 L/min, respectively. The experimental and simulated errors are within 10%; thus, the accuracy of the simulation is verified. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

23. STUDY ON THE CHARACTERISTICS OF MICRO-PITTING DAMAGES FOR HIGH POWER DENSITY GEAR TRANSMISSION

Author

LI JiQiang, LIU ZhongMing, WANG DongFei, LI YouHua, and ZHANG YanKun

Subjects
Gear transmission, Micro-pitting, Damage mechanism, Fatigue analysis, Lubrication design, Mechanical engineering and machinery, TJ1-1570, Materials of engineering and construction. Mechanics of materials, TA401-492
Abstract

the formation mechanism of micro-pitting had been described in this paper,the action relationships of various influencing factors had also been analyzed,and the three main action mechanisms of local contact stress,film thickness ratio of lubricating oil and micro mechanical properties for materials had been regularized.Based on elastohydrodynamic lubrication theory and previous test results for Newtonian fluids,the reasonable calculation method of film thickness for lubricating oils had been determined.Through comparative calculations,the specific effects of oil types,lubricating oil temperatures,transmission torques and gear speeds on the oil film thickness had been analyzed.Combined with the micro-pitting design of gear transmission,feasible ideas and operational methods for lubrication improvements and optimizations had also been given.Finally,through the experimental designs,the correctness of the theoretical analysis had been verified,and the basic data for the anti micro-pitting design of the combination for gears of 18 CrNiMo7-6（material）+ISO-VG220（Lubricating oil） had been goted.

Published
2020
Full Text
View/download PDF

24. Calculation and Analysis of Equilibrium Position of Aerostatic Bearings Based on Bivariate Interpolation Method.

Author

Li, Shuai, Huang, Yafu, Yu, Hechun, Wang, Wenbo, Zhang, Guoqing, Kou, Xinjun, Zhang, Suxiang, and Li, Youhua

Subjects
INITIAL value problems, FINITE difference method, INTERPOLATION, EQUILIBRIUM
Abstract

The solution of equilibrium positions is a critical component in the calculation of the dynamic characteristic coefficients of aerostatic bearings. The movement of the rotor in one direction leads to bidirectional variations in the air film force, resulting in low efficiency when using conventional calculation methods. It can even lead to iterative divergence if the initial value is improperly selected. This study concentrates on the orifice throttling aerostatic bearings and proposes a novel method called the bivariate interpolation method (BIM) to calculate the equilibrium position. The equilibrium equation for the rotor under the combined influence of air film forces, gravity, and external loads is established. A calculation program based on the finite difference method is developed to determine the equilibrium position. The process of solving the equilibrium position and the convergence is compared with the secant method and the search method. Furthermore, the variation trend of the equilibrium position and stiffness when the external loads changes are studied based on the BIM. Finally, the correctness of the BIM to solve the equilibrium position is proved by comparing it with the experiment results. The calculation results indicate that the BIM successfully resolves the problem of initial value selection and exhibits superior computational efficiency and accuracy. The equilibrium position initially moves away from the direction of the external load as the load increases, and then this gradually approaches the load direction. The main stiffness increases with increases in the external load, while the variation in cross stiffness depends on the direction of the external load. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

27. Discussion and Deliberation on the Latest ISO 6336 Standard

Author

Liu Zhongming Li Youhua Zhang Zhihong Ding Wei Chen Chao

Subjects
Gear strength, Mechanical engineering and machinery, TJ1-1570
Abstract

Compared with ISO 6336:1996 edition，there are great changes in the latest edition of ISO/FDIS 6336-1 to 3：2019, in particular, the calculation of the helix angle factor for helical gear has a subversive change. In order to facilitate the reader to understand the changes of the new ISO 6336, the main changes in the content of different editions of ISO 6336 are introduced, the influences of the helix angle factors and single pair tooth contact factor on safety factor before and after modification are analyzed, and the influence of helix angle on the calculation results of contact strength and bending strength are discussed here with examples.

Published
2019
Full Text
View/download PDF

28. Calculation Analysis and Test Verification of Transmission Efficiency of Rack Test Device for the Three Gorges Ship Lift

Author

Li Youhua, Li Quancai, Liu Zhongming, Yu Feipeng, and Liu Yuechan

Subjects
Ship lift, Rack, Gearbox, Transmission efficiency, Mechanical engineering and machinery, TJ1-1570
Abstract

The rack test device for the Three Gorges ship lift is mainly composed of four parts: driving gearbox,driving gear-rack,loading gear-rack and loading gearbox. In order to better calculate and analyze the transmission efficiency of the rack test device,through the establishing gearbox transmission efficiency calculation model,transmission efficiencies of the four parts and the transmission efficiency and the power consumption of the total whole unit are calculated respectively,the calculation results are contrastive analyzed by the input speed and torque,output speed and torque of the test data,the accuracy of calculation model and test data is verified,the research basis of transmission efficiency of gearbox and gear rack for low speed and load duty is established.

Published
2018
Full Text
View/download PDF

29. Experimental Study on the Influences of Backlash and Elastic Deformation on the Bidirectional Operation Characteristic of Gearbox

Author

Li Youhua, Liu Zhongming, Zhang Zhihong, Li Jiqiang, Liu Yuechan, and Yu Feipeng

Subjects
Meshing backlash, Torsion deformation, Output speed, Gearbox, Mechanical engineering and machinery, TJ1-1570
Abstract

The gearbox of rack test for the Three Gorges ship lift is low speed heavy duty four stages transmission,and need frequent fast reversing operation in the process of test,due to the influence of the gear meshing backlash,elastic torsional deformation of shaft and elastic deformation of gear tooth and other factors,the operation performance of gearbox will be changed during reversing process,especially the output speed and torque. In order to analyze the influence on the transmission performance of gearbox,the calculation models of the influence of the gear meshing backlash and the elastic torsional deformation of the shaft on the rotational speed are setup respectively,the output speed of the gearbox in the reversing process is calculated and compared with the output speed of gearbox measured by experiment,which verifies the accuracy of the calculation model and the test data.

Published
2018
Full Text
View/download PDF

30. Analysis and Test of the Transmission Efficiency of Low Speed and Heavy Duty Gearbox

Author

Li Youhua, Li Quancai, and Liu Zhongming

Subjects
Gearbox, Efficiency calculation, Data processing, Mechanical engineering and machinery, TJ1-1570
Abstract

Through the analysis of the influence factors on transmission efficiency of gearbox,a theoretical analysis model of gearbox transmission efficiency is set up from the power consumption,and the influences of gear meshing loss,bearing friction loss,oil loss,oil seal friction loss and windage loss on gearbox transmission efficiency are considered. The analysis model is validated according to the test data of a low speed and heavy duty gearbox,and the influences of speed and torque on the transmission efficiency of gearbox are discussed,the changing curves of gear efficiency with the rotational speed and torque are obtained,the foundation for further research on the efficiency of the gearbox is laid.

Published
2017
Full Text
View/download PDF

42. An investigation of haptic modelling for oral and maxillofacial surgical training and planning

Author

Yan Xiu-Tian, Smart Susie, Drikakis Dimitris, Ayoub Ashraf, Li Mutian, and Li Youhua

Subjects
Focus (computing), medicine.medical_specialty, Virtual reality simulator, Computer science, Model based simulation, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Surgical training, GeneralLiterature_MISCELLANEOUS, stomatognathic diseases, stomatognathic system, Human–computer interaction, Dental surgery, TA164, medicine, Sensibility, Haptic technology, ComputingMethodologies_COMPUTERGRAPHICS
Abstract

This research investigates a haptic modelling approach where high resolutions are required for sensibility of force feedback in a target application - dental surgical operations. In particular the research focus is on maxillofacial deformity operations. The main aim of the research is to increase the realism of a computer model based simulation system that allows dental students and surgeons to feel like as if they were carrying out a real dental surgery procedure. A haptic framework has been designed and implemented to demonstrate its suitability to achieve the above aim. A generic set of jaw bone models have been developed and validated by collaborating surgeon from Glasgow University Dental Hospital & School. The model can be customized and obtained for each specific patient. A chosen model can be used to control the force feedback generated by a haptic device in order to give a realistic force feedback representation and experience for a user. This meets the requirements of the targeted dental operations. The simulation model is constructed based on force calculation model, which include the consideration of mechanical properties of bones, cutting tools used and region of cut. This haptical jaw bone model has been generalised based on a contact mechanic model Hertz's equation and is used as the driving haptical model for the study. Using a haptic device as a controller, a user can perform relevant operations, such as cutting procedure and manipulating bone segments from the virtual jaw bone model. The paper describes a frame work for a haptic assisted surgical plan (HASP) for surgical planning of dentofacial deformities. The haptic system generates the corresponding force feedback to a user as in a real world operation. It is planed that validation and feedback experiments will be conducted with dental students and surgeons to assess the effectiveness of the system in providing assistance for oral and maxillofacial surgical planning and training.

Published
2021

43. VST3D-Net:Video-Based Spatio-Temporal Network for 3D Shape Reconstruction from a Video

Author

Guanglun Zhang, Li Youhua, Yang Lu, and Yang Jinglun

Subjects
Computer science, business.industry, Feature extraction, Frame (networking), ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Iterative reconstruction, Object (computer science), ENCODE, Feature (computer vision), Computer vision, Artificial intelligence, business, Subnetwork, Pose
Abstract

In this paper, we propose the Video-based Spatio-Temporal 3D Network (VST3D-Net), which is a novel learning approach of viewpoint-invariant 3D shape reconstruction from monocular video. In our VST3D-Net, a spatial feature extraction subnetwork is designed to encode the local and global spatial relationships of the object in the image. The extracted latent spatial features have implicitly embedded both shape and pose information. Although a single view can also be used to recover a 3D shape, more rich shape information of the dynamic object can be explored and leveraged from video frames. To generate the viewpoint-free 3D shape, we design a temporal correlation feature extractor. It handles the temporal consistency of the shape and pose of the moving object simultaneously. Therefore, both the canonical 3D shape and the corresponding pose at different frame are recovered by the network. We validate our approach on the ShapeNet-based video dataset and ApolloCar3D dataset. The experimental results show the proposed VST3D-Net can outperform the state-of-the-art approaches both in accuracy and efficiency.

Published
2020

49. Risk Assessment and Service Benefit Evaluation of Forestry Meteorological Disasters in Guangxi

Author

HUANG, Shuyan, LI, Yanjun, LI, Youhua, GAO, Shan, and LI, Wei

Subjects
Forestry, Risk census, Meteorological service, Benefit evaluation, Agribusiness
Abstract

In order to strengthen the construction of capacity of forestry meteorological services and improve the quality of forestry meteorological services, the national forestry meteorological disaster risk survey and service benefit assessment were carried out from November 2016 to May 2017. A total of 196 risk sites in Guangxi were collected by means of questionnaire survey, expert evaluation and field survey. The results showed that fire, pest and wind hazards are the main types of risk points in Guangxi. The survey showed that there were 77 monitoring stations around the disaster risk sites belonging to the Forestry Sector, which mainly monitor fire hazards, while other disasters had no monitoring point yet. Early warning facilities were still constantly improving. The meteorological disasters that affecting forestry production were drought, gale, and heavy rainfall, and the main meteorological hazard factors were temperature, wind speed, and precipitation. Forest fire prevention, forestry resources development and utilization, forestation and tending demand for meteorological services were relatively large. It was hoped that forestry meteorological service products would be obtained through telephone, SMS and early warning systems. In addition, experts gave an objective evaluation of the contribution rate of meteorological services in Guangxi. The contribution rate of forest meteorological services in the whole region was 8.41%. The scale of Guangxi’s total forestry output value was 31.478 billion yuan in 2016. According to the calculation of this value, the service benefit value had reached 2.647 billion yuan. In conclusion, strengthening the cooperation in forestry meteorological monitoring, technology development, and emergency response, and further improving forestry meteorological services and effective reducing forestry disaster losses are the top priorities of the meteorological department.

Published
2019
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results